SALAMANDRA 56(1): 66–70 SALAMANDRA 15 February 2020 ISSN 0036–3375 Correspondence German Journal of Herpetology

Correspondence

Aspects of the reproductive biology of vandijki with a description of neonates

Steven G. Platt1, Myo Min Win1, Kalyar Platt2, Nathan A. Haislip3 & Thomas R. Rainwater4

1) Wildlife Conservation Society – Myanmar Program, No. 12, Nanrattaw St., Kamayut Township, Yangon, Myanmar 2) Survival Alliance – Myanmar Program, No. 12, Nanrattaw St., Kamayut Township, Yangon, Myanmar 3) Turtle Survival Alliance – Turtle Survival Center, 1030 Jenkins Road, Suite D, Charleston, South Carolina 29407, USA 4) Tom Yawkey Wildlife Center & Belle W. Baruch Institute of Coastal Ecology and Forest Science, Clemson University, P.O. Box 596, Georgetown, South Carolina 29442, USA

Corresponding author: Thomas R. Rainwater, e-mail: [email protected]

Manuscript received: 23 July 2019 Accepted: 11 November 2019 by Edgar Lehr

The Burmese Narrow-headed Softshell Turtle (Chitra endangered (Dayton 2003). We here report on as- vandijki McCord & Pritchard, 2003) is a large (straight- pects of the reproductive ecology (phenology, nesting hab- line carapace length [CL] to ca. 1000 mm) softshell turtle itat, clutch and attributes, and incubation period) of endemic to Myanmar where confirmed records are avail- C. vandijki in the upper Chindwin River of Myanmar, and able from the Ayeyarwady-Chindwin River system, and provide the first published description of neonates with ac- lower Sittang and Thanlwin rivers (Kuchling et al. 2004, companying photographs. Platt et al. 2005, 2014, 2018). Chitra vandijki is subject to The Chindwin River arises in the Hukaung Valley of intense subsistence exploitation by rural villagers, large northern Myanmar and flows southward for about 1200 numbers are taken for illegal export to wildlife markets km before debouching into the Ayeyarwady River near in southern China, are widely collected for domestic Pakokku (Gresswell & Huxley 1965). Much of the up- consumption, and habitat degradation due to gold-mining per (upstream from Homalin) Chindwin River is charac- and conversion of sandbanks to seasonal agricultural fields terized by wide meanders with extensive sandbanks and is occurring on many rivers (Kuchling et al., 2004, Platt point bars, although some stretches are confined to a deep- et al. 2014, 2017b). Consequently, populations of C. vandijki ly incised channel with small sandbanks nestled in pockets throughout Myanmar have declined precipitously (Platt eroded from the bedrock (Platt et al. 2013, 2017a). The et al. 2014), even within some protected areas (Platt et al. regional climate is tropical monsoonal with an annual wet 2017b). As a result, C. vandijki is now in the process of be- season extending from late May–early June through late ing classified as Critically Endangered on the IUCN Redlist September, followed by a dry season from mid-October to and ranked among the 25 most endangered chelonians in May (Terra 1944, Platt et al. 2013). River levels reflect the the world (Stanford et al. 2018). monsoonal cycle with peak water levels occurring during Despite this parlous conservation situation, virtually the mid- to late wet season (Platt et al. 2013, 2017a). nothing is known concerning the ecology of C. vandij­ Our observations of C. vandijki reproduction were ki (Platt et al. 2014). This is certainly the case with re- made in conjunction with a conservation project focused gards to reproduction (Platt et al. 2014); with the excep- on the critically endangered (Bata­ tion of a single clutch of eggs obtained from a villager and gur trivittata [Duméril & Bibron, 1835]) in the Chindwin described by Platt et al. (2018), information on the re- River (reviewed by Platt & Platt 2018). We obtained productive ecology of C. vandijki is non-existent. Further- clutches of C. vandijki eggs 3–24 hours after being notified more, descriptions and photographs of neonate C. vandijki by Community Conservation Cadres (CCC) employed to are unavailable in the literature (Platt et al. 2014). This monitor sandbanks along the river for signs (e.g., track- dearth of information is lamentable because such basic ways) of turtle nesting activity. At each nest, we first meas- natural history data are a necessary prerequisite for design- ured the distance from the approximate center of the nest ing and implementing effective conservation strategies for to the river. We then began excavating the nest until reach- © 2020 Deutsche Gesellschaft für Herpetologie und Terrarienkunde e.V. (DGHT), Mannheim, Germany Open66 access at http://www.salamandra-journal.com Correspondence

Table 1. Laying and hatching dates, distance to water, and clutch attributes of three Chitra vandijki nests found on sandbanks along the upper Chindwin River, Sagaing Region, Myanmar in 2018.

Egg diameter (mm) Nest Laying date Hatching date To river Clutch Mean ± 1SD Range (m) size 1 21 July 2018 30 September–1 October 2018 21 76 28.3±0.4 27.1–29.6 2 15 August 2018 29 October–5 November 2018 31 102 29.5±1.0 27–36.6 3 7 September 2018 Undetermined (see text) 20 89 26.0±3.2 20.1–30.5 ing the uppermost eggs in the clutch and measured the vated 20–31 m from the edge of the river. Although posi- vertical distance from the sand surface to the eggs. We re- tioned on level (slope < 5°) microsites near the highest part moved the clutch (taking care to maintain the upright ori- of the sandbank, a slight rise in river levels (ca. 30–45 cm) entation of each egg), counted the eggs, and measured the would probably have inundated all three nests. Female vertical distance from the surface to the floor of the nest B. trivittata have used the same sandbanks for nesting dur- cavity. We measured the diameter of each egg using dial ing February and March (peak months of dry season) in calipers (± 0.1 mm), packed the eggs in a sand-filled plastic previous years (Platt et al. 2017a). box and transported them by boat to a secure husbandry The three clutches ofC. vandijki eggs that we found were facility at Limpha Village (25°48’19’’ N, 95°31’44’’ E, 132 m deposited from 21 July to 7 September 2018 (Table 1). We above sea level). Clutches were incubated under ambient also purchased 15 eggs in the Khamti Market on 16 August conditions in either a sand-filled Styrofoam Box (1 clutch) 2018. Because turtle eggs are unlikely to remain unsold for or artificial sand bed surrounded by a tightly woven bam- more than a few days, we assume this clutch was depos- boo fence lined with plastic sheeting to prevent the escape ited around 14–15 August 2018. Previously, we obtained a of hatchlings (two clutches). clutch of 56 C. vandijki eggs from a villager in Padumone We measured neonates within 24 hours of emerging (26°00’55’’ N, 95°52’09’’ E, 183 m above sea level) on 25 Au- from the nest and if the laying date was known, calculated gust 2016 that was harvested 1–2 days earlier (Platt et al. the incubation period. Using dial calipers, we measured (to 2018). Collectively, these data indicate that egg-laying by the nearest 0.1 mm) CL, maximum carapace width (CW), C. vandijki occurs during a two-month period from mid- mid-line plastron length (PL), and shell depth (SD; from July through early September, a time that coincides with plastron to highest point of carapace) of each neonate. Our near-maximal river levels of the late wet season (Fig. 1). measurements include the pliable margins of the bony car- Given this phenology, C. vandijki nests are probably at risk apace (Leathery Carapace Length of Pritchard 2001) and of being lost to flooding during most years. plastron. Neonates were maintained in captivity for about We calculated mean clutch size for C. vandijki based on six months (head-started) and then released into the wild. the three nests found along the Chindwin River in 2018 We also searched for turtle eggs during infrequent visits (Table 1) and another clutch of 56 eggs from Padumone to an urban market in Khamti (25°59’48’’ N, 95°42’06’’ E, (Platt et al. 2018). Eggs purchased in the Khamti Market 144 m above sea level). Eggs purchased in the market were were not included in our analysis because turtle eggs are returned to Limpha, measured, and incubated as described sold individually and those we obtained were almost cer- above. River level and air temperature data were obtained tainly part of a larger clutch from which eggs had already from the Department of Meteorology and Hydrology, been removed. The mean size of the four complete clutches Ministry of Transport and Communication in Khamati in our sample was 80.7 ± 19.6 eggs (range = 56–102 eggs). and Homalin, respectively (air temperature is not recorded Chitra vandijki eggs are spherical in shape with a leathery at Khamti station). Mean values are presented as ± 1SD. shell that becomes increasingly hardened as eggs dry dur- We found three C. vandijki nests on two sandbanks ing incubation. The mean egg diameter for our complete along the western shore of the Chindwin River near Sein sample (including eggs from Khamti Market) was 28.2 ± Naing Village (25°15’39’’ N, 95°10’12’’ E; 140 m above sea 2.3 mm (n = 338; range = 20.1–36.6 mm). level) during 2018 (Table 1). Conspicuous trackways left We incubated 282 C. vandijki eggs (267 collected from by nesting female led from the river to each nest. the wild and 15 purchased in market) of which 150 (53.1%) The sandbanks used for nesting by female C. vandijki are successfully hatched. None of the eggs we obtained in the adjacent to deep holes in the riverbed, and composed Khamti Market hatched, most likely as the result of mis- largely of medium to fine-grained metamorphic sands, a handling prior to our purchase. Confusion as to when well-drained, porous substrate that permits gaseous diffu- hatchlings first emerged from eggs collected on 7 Septem- sion and heats rapidly, thereby enhancing embryo growth ber 2018 precluded an accurate determination of the incu- (Platt et al. 2017a). Vegetative cover was absent at the nest bation period for that clutch. Otherwise, eggs deposited in sites, which were exposed to direct sunlight for most of the July and August hatched in late September, and late Oc- day. Nests were deep holes of approximately uniform depth tober–early November after incubation periods ranging (ca. 300 mm to eggs and 450 mm to bottom of hole) exca- from 71–72 and 75–81 days, respectively (Table 1). Simi-

67 Correspondence larly, Platt et al. (2018) reported an incubation period Table 2. Morphometric measurements of 150 neonate Burmese of 81 days for a clutch of C. vandijki eggs collected in Au- Narrow-headed Softshell Turtles (Chitra vandijki) hatched from gust 2016. Air temperatures were relatively stable during eggs collected along the upper Chindwin River, Sagaing Region, the incubation period (21 July through 5 November 2018); Myanmar. minimum and maximum air temperatures averaged 23.9 ± 2.0°C and 31.6 ± 2.7 °C, respectively. For the most part, riv- Attribute Mean ± 1SD (mm) Range (mm) er levels gradually declined during the incubation period Carapace length 35.2 ± 3.5 27.3–41.0 and were approaching the seasonal minima when hatch- Carapace width 33.9 ± 2.6 27.5–38.6 lings emerged from the nests (Fig. 1). Plastron length 30.3 ± 1.6 24.9–34.6 Morphometric data were obtained from 150 neonate C. vandijki that emerged from eggs we collected along the Shell depth 9.4 ± 1.0 7.6–13.2 Chindwin River (Table 2). Assuming an upper asymptot- ic CL of 1000 mm for adult C. vandijki (Pritchard 2001, ied or partially buried just beneath the substrate, and wait Platt et al. 2014), the mean CL of neonates is 3.5% of the for prey (e.g., small fish). As such, the cryptic body colora- maximum adult body size. Neonate C. vandijki are char- tion of neonates provides concealment from prey and also acterized by a brown-green carapace covered with lighter lessens the chances of detection (especially when neonates (green-olive), irregular reticulations (Fig. 2a). The asym- are partially buried in the substrate or swimming) by pred- metrical dark vertical streaks over the costal bones on the ators that forage in shallow water (e.g., wading birds). Sim- carapace of large juveniles and adults are absent in ne- ilar to other hatchling trionychids (Muller 1921, Webb onates. As in the adults, distinct longitudinal stripes with 1962; Greenbaum & Carr 2002), a caruncle (egg-tooth) is a V-shaped mark are obvious on the head and neck of ne- present on the maxilla immediately ventral to the probos- onates and dark splotches surround the eyes and proboscis cis of neonate C. vandijki (Fig. 2b). The caruncle was still (Fig. 2b). The legs and feet are green-brown with lighter present on some neonates as late as mid-December 2018, reticulations consistent with similar markings on the cara- which is somewhat surprising given that others report rel- pace. The plastron of neonates is uniform whitish-grey and atively brief retention periods of ≤ 7 days (Muller 1921, the underside of the carapace margin is greyish. Our ob- Lovich et al. 2014). We speculate the lengthy retention of servations in captivity suggest neonate C. vandijki are am- the caruncle is an artefact of captivity, perhaps influenced bush predators, which inhabit shallow water, remain bur- by the nutritional status of the neonates.

Figure 1. Laying and hatching dates for Chitra vandijki eggs in relation to water levels in the Chindwin River, Myanmar (2018). Nest number corresponds to Table 1. 1 = Laying date (A) and hatching date (B) for eggs in Nest 1. 2 = Eggs obtained in Khamti Market. 3 = Laying date (A) and hatching date (B) for eggs in Nest 2. 4 = Probable laying date (A) and hatching date (B) for clutch reported by Platt et al. (2018). 5 = Clutch in Nest 3 deposited (hatching date uncertain; see text). River level data courtesy Department of Meteorology and Hydrology, Ministry of Transport and Communication, Government of Myanmar.

68 Correspondence

In summary, our observations along the Chindwin Riv- sandbanks as critical nesting habitat for endangered chelo- er indicate that C. vandijki deposits eggs late in the wet sea- nians (e.g., C. vandijki, N. formosa, and B. trivittata) in the son when river levels are near the seasonal maximum and upper Chindwin and other rivers in Myanmar where these hatching occurs in the late-wet and early-dry seasons as riv- species occur. Successful conservation therefore depends er levels are falling. This phenological pattern differs from in part on effectively protecting these habitats from con- that of other sympatric river turtles (e.g., trivittata version to other uses, such as seasonal agricultural fields and formosa [Gray, 1869]) in central and north- and sand-mining. ern Myanmar, which deposit eggs during the mid-dry sea- son with hatching timed to coincide with rising river lev- els at the onset of the wet season (Platt et al. 2016, 2017a, Acknowledgements 2018). Our earlier speculation that C. vandijki followed this pattern (Platt et al., 2014) was obviously in err, and We thank the Ministry of Environmental Conservation and For- estry for granting us permission to conduct research and sup- here stands corrected. Our observations also confirm that porting our biodiversity conservation efforts in Myanmar. This like other congeners, C. vandijki produces large clutches of project was made possible by generous grants from A. Sabin and relatively small eggs (Nutaphand 1979, Pritchard 2001, the Andrew Sabin Family Foundation, and the Critical Ecosystem Kitimasak et al. 2005). Finally, our observations (Platt Partnership Fund. We are grateful for administrative assistance et al. 2017a; this study) and those of Kuchling et al. (2004, and support of Than Myint, Saw Htun, C. Poole, and R. Hud- 2006), further high-light the importance of undisturbed son. We thank the Department of Meteorology and Hydrology, Ministry of Transport and Communication (Khamati and Homa- lin) for providing us with river level and air temperature data, San Phay, Zaw Naing Oo, Naing Win Aung, Tun Win Zaw, A Khaung Khat Zaw, and Sie Mue Htaung for field assistance, and J. Iverson, G. Kuchling, and L. Medlock for literature ref- erences and thoughtful commentary. Our manuscript benefited from the reviews of P. Wagner and P. P. van Dijk. This paper represents Technical Contribution Number 6773 of the Clemson University Experiment Station.

References

Dayton, P. K. (2003): The importance of the natural sciences to conservation. – American Naturalist, 162: 1–13. Greenbaum, E. & J. L. Carr (2002): Staging criteria for embryos of the , spinifera (Testudines: Tri- onychidae). – Journal of Morphology, 254: 272–291. Gresswell, R. K. & A. Huxley (1965): Standard encyclopedia of B the world’s rivers and lakes. – G.P. Putnam’s Sons, NY. Kitimasak, W., K. Thirakhupt, S. Boonyaratpalin, & D. L. Moll (2005): Distribution and population status of the nar- row-headed softshell turtle Chitra spp. in Thailand. – Natural History Bulletin of Chulalongkorn University, 5: 31–42. Kuchling, G., Win Ko Ko, Sein Aung Min, Tint Lwin, Khin Myo Myo, Thin Thin Khaing (1), Thin Thin Khaing (2), Win Mar Mar & Ni Ni Win (2006): Two remnant popu- lations of the roofed turtle Kachuga trivittata in the upper Ayeyar­wady River system, Myanmar. – Oryx, 40: 176–182. Kuchling, G., Win Ko Ko, Tint Lwin, Sein Aung Min, Khin Myo Myo, Thin Thin Khaing (2), Win Mar Mar & Thin Thin Khaing (1) (2004): The softshell turtles and their exploi- tation at the upper Chindwin River, Myanmar: range exten- sions for cartilaginea, Chitra vandijki, and Nilssonia formosa. – Salamandra, 40: 281–296. Lovich, J. E., C. H. Ernst, E. M. Ernst & J. L. Riley (2014): A 21-year study of seasonal and interspecific variation of hatch- ling emergence in a Nearctic freshwater turtle community: Figure 2. Neonate Chitra vandijki hatched from eggs collected To overwinter or not to overwinter? – Herpetological Mono- along the upper Chindwin River in Myanmar. (A) View of cara- graphs, 28: 93–109. pace and head. (B) Caruncle below proboscis (white arrow). Pho- Muller, J. F. (1921): Notes of the habits of the soft-shell turtle tographs by N. A. Haislip. – Amyda mutica. – American Midland Naturalist, 7: 180–184.

69 Correspondence

Nutaphand, W. (1979): The turtles of Thailand. – Siamfarm Zoo- logical Gardens, Bangkok. Platt, S. G. & K. Platt (2018): Tide slowly turns in the fight to save Myanmar’s critically endangered turtles. – Turtle Surviv- al, 2018: 41–48. Platt, S. G., K. Platt, Khin Myo Myo, Kyaw Moe, Me Me Soe, Thet Zaw Naing, Naing Lin & T. R. Rainwater (2013): Noteworthy records of chelonians from the Chindwin River basin and Naga Hills of western Myanmar. – Herpetological Conservation & Biology, 8: 335–350. Platt, S. G., K. Platt, Win Ko Ko & T. R. Rainwater (2014): Chitra vandijki McCord and Pritchard 2003 – Burmese Nar- row-headed Softshell Turtle. – Chelonian Research Mono- graphs, 5: 74.1–74.7. Platt, S. G., E. A. Measures, D. M. Rohr, K. Platt & T. R. Rainwater (2017a): Batagur trivittata (Burmese Roofed Tur- tle). Nesting site and substrate. – Herpetological Review, 48: 420–422. Platt, S. G., Tint Lwin, Aung Lin Htet, K. Platt, P. P. van Dijk & T. R. Rainwater (2016): Nilssonia formosa (Burmese Pea- cock Softshell Turtle). Morphometrics, coloration, and photo- graph of hatchling. – Herpetological Review, 47: 125–126. Platt, S. G., Tint Lwin, Naing Win, Htay Lin Aung, K. Platt & T. R. Rainwater (2017b): An interview-based survey to determine the of softshell turtles (Repti­ lia: ) in the Irrawaddy Dolphin Protected Area, Myanmar. – Journal of Threatened Taxa,9 : 10998–11008. Platt, S. G., Win Ko Ko, Lay Lay Khaing, Khin Myo Myo, Tint Lwin, Thanda Swe, Kalyar & T. R. Rainwater (2005): Noteworthy records and exploitation of chelonians from the Ayeyarwady, Chindwin, and Dokhtawady rivers, Myanmar. – Chelonian Conservation & Biology, 4: 942–948. Platt, S. G., G. R. Zug, K. Platt, Win Ko Ko, Khin Myo Myo, Me Me Soe, Tint Lwin, Myo Min Win, Swann Htet Naing Aung, Nay Win Kyaw, Htun Thu, Kyaw Thu Zaw Wint, P. P. van Dijk, B. D. Horne & T. R. Rainwater (2018): Field re- cords of turtles, snakes and lizards in Myanmar (2009–2017) with natural history observations and notes on folk herpeto- logical knowledge. – Natural History Bulletin of the Siam So- ciety, 63: 67–114. Pritchard, P. C. H. (2001): Observations on body size, sympat- ry, and niche divergence in softshell turtles (Trionychidae). – Chelonian Conservation & Biology, 4: 5–27. Stanford, C. B., A. G. J. Rhodin, P. P. van Dijk, B. D. Horne, T. Blanck, E. V. Goode, R. Hudson, R. A. Mittermeier, A. Currylow, C. Eisemberg, M. Frankel, A. Georges, P. M. Gibbons, J. O. Juvik, G. Kuchling, L. Luiselli, S. Haitao, S. Singh & A. Walde (2018): Turtles in trouble: The World’s 25+ most endangered and freshwater turtles – 2018. Turtle Conservation Coalition, Ojai, CA. Terra, H. de. (1944): Component geographic factors of the natu- ral regions of Burma. – Annals Association of American Ge- ographers, 34: 67–96. Webb, R. G. (1962): North American recent soft-shelled turtles (Family Trionychidae). – University of Kansas, Publications of Natural History Museum, 13: 429–611.

70